Emeritus Prof. Computing + Information physics, Promise Theory, CFEngine, author of
@SmartSpaceTime2
++, tech/leadership advisor ChiTek-i, varied music composer.
Weekend Thread - Promise Theory Intro (PT)
----
1. PT is a general theory about the mechanics of abstract "agents" and their "intended" behaviours, i.e. processes. It was originally proposed as a model for Computer Networks, but has since been applied to many different scenarios.
@nattyover
On behalf of all of us who purposely left academia and happily, I hope the "no academic affiliation" slur is seen as evidence that academic affiliation is no criterion by which to judge intelligence or research capability--it only reflects a snobbery in elites and sci-journalism.
The main key to understanding any problem is to find the optimum naming convention, notation, dynamic and semantic coordinates. It's worth taking the time to get it wrong then make it right. Iterations!
1. This is a thread about Quantum Mechanics and how we interpret it. The way QM is presented bothers me from time to time (at least when I have time to think about it). Indeed, time plays the key role in the whole issue.
I am constantly told that no one wants the kind of work I do, or the things I write. It's too smart, too broad, too this too that. I must be mad to take on the challenges. Each one is a world of hurt. But I persist because I believe I can help a tiny few to go further. Sigh.
25 years ago, I co-wrote one of those oddball theoretical papers on Dirac fermions confined to curved surfaces, as a loose-end curiosity of curved spacetime. Today this paper is being cited more than almost any other I wrote in physics as it suddenly applies to nano-technology!
I'm delighted to announce that, after 6 years, a much improved second edition of Promise Theory book is now out! Along with it an important application of PT to the theory of money! View at
Cc
@jeffsussna
@botchagalupe
@DanielMezick
Scaling in a nutshell:
- When things grow big, they can't always sustain what they did well when they were small.
- When things become many, they can't easily maintain the coherence afforded as a few.
5 years seeking certainty:
Most people / organizations have no patience for solving hard problems. Lacking a vision and not trusting, they give up long before progress can be made. They offer no mandate for innovation so it's up to independent lone wolves to have the stamina.
For everyone who thinks they know statistics, or has caught themselves 'splaining "correlation is not causation" - Pearl's long awaited readable book (his earlier books being not). Why causation exists and why statistics can't do it. Recommended.
1. This is one of several threads about databases and distributed systems, and why these are still a headache for enterprises after so many years. Part 1, basics. N=2^5
Finally drawing a line under this paper.
Locality, Statefulness, and Causality in Distributed Information Systems
Thank you to all who reviewed it for me and improved it as a result. Now I call on everyone to use it to start an informed discussion! ❤️
1/2..
To my engineering friends -- I would like your help. I've been compiling this book (especially volume 2) of theory of systems for years, and here is finally a rough draft, more or less complete.
I try not to use the Internet as my psychiatrist, but on a dark day I let down my guard - and I am blown away by the many kind responses to a throwaway remark. Thank you to everyone who sent their support. That certainly lifted my spirits- you truly learn who your friends are. 🙈
"A landmark in the development of our craft" - five years since this book was released, with foreword by
@adrianco
- - my first book in the hope to inspire young and open minded
12.We often note that quantum processes can be in superpositions, and say that this is fundamentally quantum, but it isn't. It's really a question of sampling rates. It's known that "quantum uncertainty" is actually Nyquist sampling rate uncertainty in a Fourier analysis of waves
@nattyover
Note however, "brilliant laypeople" is the second remark that I would think is a slur. Just because we are not at a university doesn't make us laypersons.
Warms my heart a little to know that two of the largest internet giants still run all their systems on
@cfengine
-- hundreds of thousands of computers cooperating to reach a managed equilibrium.
"Getting close to the time now" - what a muddle of space and time concepts we use in everyday language! I'll be explaining why, and much more ... my new book is in its final stages.
New draft paper on "Locality, Statefulness, and Causality in Distributed Information Systems (Concerning the Scale Dependence Of System Promises)" now available for comment and peer review at. I welcome feedback. Applicable to web, data pipelines, ++ etc
My delightful partner has been practising her English idioms. She confronted me with a sly look this morning, and announced: "Well, someone's village is missing their idiot!" Then burst out laughing.
Using Promise Theory to solve the distributed consensus problem- another in my series of critical essays about IT practice. It's another of those problems we make harder than it needs to be out of habit...there are better answers if we care to look.
Today I'm starting a new series about the concepts of Semantic Spacetime applied to data analytics. I'll be showing practical examples and code using
#golang
and
#arangodb
. Stay tuned after this introduction to follow the ten parts, each week!
You're not doing theoretical physics unless you're fine tuning, right? The sensitivity of discrete systems is infuriatingly unstable but my interference patterns are at least symmetricalish now. .🥴😅
Today, I am releasing the self-paced learning videos about Promise Theory Basics, based on collaboration with the for general free access. I hope this will eventually help to increase understanding of this useful methodology.
1. It's config management time, so here is a thread about configuration languages in IT. Config languages are not unique to IT, they exist in all areas of science and engineering to describe layout or "state".
1. Segal's Law: "If you have one clock, you know what time it is. If you have more than one, you're not sure." An eyebrow later, everyone gets this. Yet, in distributed computing, we don't always stop to think what it means for time sensitive operations.
This is now pending release on Amazon, over 1100 pages of systems modelling experience, freely available thanks to my work with Countdown has begun....
As a result of a couple of recent personal tragedies, a close friend prompted me to write down some personal stories from my life. It turns out I did a few things. I wonder if that's something for public or private view...
The first complete rough cut of my Treatise on Systems, now finally complete at
This has been exhausting! Now begins the long process of QA and optimization. Reviewers welcome!
I spent 20 years now trying to bridge the world between academia and industry, and academia is by far the least able to step into that bridge. Filled with false pride and self importance it only wants industry's money, but it's not willing to adapt to deserve it.
Thanks to everyone who offered to look at my draft paper. A bit overwhelmed by the interest. But ok - great. I going to put the draft here for all interested. Please email comments (see my contact page) and there will be a cut off in a couple of weeks.
Throughout my career, my life, I never judged people by their family background, money, or school. I'm pretty easy going, and will talk to anyone with the same basic respect. But why is the world just dominated by awful snobs?
My beloved to me: "You need to start playing more dumb so that people will like you better. You don't have to solve every problem, just mumble something nice to them instead of being so smart all the time!" Me: 🤖
Years ago, I wrote about the misunderstandings around push vs pull in load balancing, with a postdoc who came to study PT. It's sort of the basis for CDN. Suddenly, this work has been getting a lot of attention for some reason.
The separation of "fast and slow" variables remains the most useful approach to the description of processes, regardless of model, continuous/discrete, etc. And when there is no natural separation, you can start to talk about the meaning of multiscale phenomena and complexity.